Under and Overmentalizing in Personality Disorders: A Principal Component Analysis of Nonadaptive Personality and the Movie Assessment of Social Cognition.

INTRODUCTION

This secondary analysis of quality control data assessed principal components of personality dysfunction and their relationship to mentalizing in a sample of treatment-seeking women with severe personality disorders.

METHODS

The Schedule for Nonadaptive and Adaptive Personality (SNAP) and the Movie for the Assessment of Social Cognition (MASC) were administered to 37 females in routine quality assessments of a specialized residential treatment program. Principal component analysis (PCA) of SNAP scores was used to determine dimensions of personality most significantly contributing to overall maladaptive personality functioning. Bootstrapped stepwise regression tested the relationship of dimensional personality indices to hypermentalizing and hypomentalizing on the MASC controlling for general psychiatric severity.

RESULTS

Four principal components (PCs) explained 71.4% of the variance in personality dysfunction, mapping onto antisocial, obsessive compulsive, borderline, and narcissistic personality constellations. The borderline and antisocial PCs were positively predictive of hypermentalizing. The obsessive-compulsive PC was positively predictive of hypomentalizing, while the antisocial PC was negatively predictive of hypomentalizing.

CONCLUSION

The study reiterates prior findings of a relationship between hypermentalizing and borderline and antisocial personality profiles. It also contributes evidence to the limited research on hypomentalizing as a clinical indicator and potential treatment target for obsessive-compulsive personality, and shows evidence of a negative relationship between antisocial personality disorder and hypomentalizing. These findings provide clinical indications for enhancing and regulating mentalizing via attention to and interpretations of internal and interpersonal events in individuals with personality disorders. Further research is needed to replicate these associations in larger, more representative clinical samples.